Visa's New AI Fraud Detection Cuts Phishing Losses by 90%
Visa has expanded its value-added services with a new fraud prevention platform that uses artificial intelligence to identify suspicious transactions in real time.
The ARIC Risk Hub, developed by Featurespace, a company Visa recently acquired, operates by creating profiles of normal customer behaviour to distinguish legitimate transactions from fraudulent activity.
Featurespace, founded at Cambridge University 15 years ago, specialises in machine learning applications for financial services.
The platform works in an open ecosystem, compatible with various payment methods beyond Visa's own network, allowing financial institutions to implement fraud prevention across their entire payment infrastructure.
Norwegian implementation demonstrates efficacy
A deployment with Eika Group, a consortium of 46 local banks in Norway, resulted in a 90% reduction in losses from phishing attacks between 2023 and 2024.
Phishing refers to deceptive communications designed to trick recipients into revealing sensitive information or transferring funds.
Eika implemented the system to address three primary challenges: increasing fraud rates targeting elderly customers, increasingly complex fraud patterns that evaded existing detection systems and operational strain from manual transaction reviews.
Jon Hagen, Chief Architect at Eika, explains how the system operates: “Featurespace's AI-powered technology learns and adapts in real-time, and profiles normal user behaviour in an efficient manner.
“As fraud patterns evolve, we will need to add more rules and maybe adjust some of the old ones. Adjustment of the rules is easy, given the powerful support of the Featurespace system.”
The technology evaluates each payment initiated through online or mobile banking in real time. Transactions deemed suspicious are flagged for investigation or blocked before completion.
Adaptive AI methodology
“As fraud patterns evolve, we will need to add more rules and maybe adjust some of the old ones. Adjustment of the rules is easy, given the powerful support of the Featurespace system"
The ARIC Risk Hub uses adaptive artificial intelligence, a form of machine learning that continuously evolves its detection capabilities based on new data inputs.
This approach differs from static rule-based systems by automatically adjusting to emerging fraud patterns without requiring constant manual updates.
The system establishes baseline behaviour profiles for each customer, enabling it to identify anomalous transactions that deviate from established patterns.
This methodology reduces false positives—legitimate transactions incorrectly flagged as fraudulent—which can impede customer experience and increase operational costs.
Antony Cahill, President of Value-Added Services at Visa, adds: “Accepting a payment is a fundamental need of any business.
“Leveraging new technology to accept payments more efficiently and securely can be what sets a business apart in today's rapidly digitalising world.
“With our new services, we're helping businesses harness data-driven insights, simplify the checkout experience, and fight fraud more effectively than ever.”
Tareq Muhmood, Head of Value-Added Services, Europe at Visa, continues: “We are thrilled to be integrating ARIC Risk Hub into our offer to businesses and financial institutions.
“We have been using AI for over 30 years to fight fraud, and now by integrating behavioural data with ARIC's cutting-edge technology we will further help our clients to stay ahead of emerging threats.”
The ARIC Risk Hub is available globally, enabling financial institutions of various sizes to implement advanced fraud prevention capabilities.
Explore the latest edition of FinTech Magazine and be part of the conversation at our global conference series, FinTech LIVE.
Discover all our upcoming events and secure your tickets today.
FinTech Magazine is a BizClik brand
